Deploying a Streamlit App on AWS, GCP, and Azure

Streamlit makes it easy to build interactive web applications in Python, but deploying them to a cloud platform like AWS, GCP, or Azure requires some setup. In this guide, we’ll cover how to deploy your Streamlit app on each of these cloud providers and discuss best practices for performance, security, and cost optimization.

1. Deploying on AWS (Amazon Web Services)

AWS provides multiple options for deploying a Streamlit app, including EC2, Elastic Beanstalk, and AWS Lambda with API Gateway. Here, we’ll focus on deploying with EC2, a flexible and widely used option.

Using EC2

1. Create an EC2 Instance:

   • Go to the AWS EC2 Console → Launch an instance.
   • Select Ubuntu 22.04 LTS (or any preferred OS).
   • Choose an instance type (e.g., t2.micro for free tier, or t3.medium for better performance).
   • Configure security group to allow inbound traffic on port 8501 (default for Streamlit).
   • Launch the instance and connect via SSH.

2. Install dependencies on EC2:

   sudo apt update && sudo apt install -y python3-pip
   pip3 install streamlit
   

3. Upload your Streamlit app (via SCP or Git):

   scp -i your-key.pem app.py ubuntu@your-ec2-ip:/home/ubuntu/
   

   Or, clone from GitHub:

   git clone https://github.com/your-repo.git
   

4. Run the Streamlit app:

   streamlit run app.py --server.port 8501 --server.address 0.0.0.0
   

5. Access your app using http://your-ec2-ip:8501/.

To keep it running persistently, use:

nohup streamlit run app.py &

Best Practices for AWS Deployment

• Use Elastic IP to ensure a fixed public IP for your instance.
• Set up Auto Scaling if you expect traffic spikes.
• Use AWS RDS instead of local databases for better reliability.
• Consider deploying with AWS Elastic Beanstalk for automated scaling and monitoring.

2. Deploying on GCP (Google Cloud Platform)

Google Cloud Platform provides multiple ways to deploy a Streamlit app, including Compute Engine, Cloud Run, and App Engine. We’ll cover Cloud Run, which is a fully managed, serverless platform.

Using Cloud Run (Serverless)

1. Install Google Cloud SDK & authenticate:

   gcloud auth login
   gcloud config set project YOUR_PROJECT_ID
   

2. Create a Dockerfile for your app:

   FROM python:3.9
   WORKDIR /app
   COPY requirements.txt ./
   RUN pip install -r requirements.txt
   COPY . .
   CMD streamlit run app.py --server.port $PORT --server.address 0.0.0.0
   

3. Build and push the Docker image:

   gcloud builds submit --tag gcr.io/YOUR_PROJECT_ID/streamlit-app
   

4. Deploy to Cloud Run:

   gcloud run deploy streamlit-app --image gcr.io/YOUR_PROJECT_ID/streamlit-app --platform managed --allow-unauthenticated --region us-central1
   

5. Get the URL and access your app!

   gcloud run services describe streamlit-app --region us-central1 --format 'value(status.url)'
   

Best Practices for GCP Deployment

• Use Google Cloud Storage to store static files efficiently.
• Optimize costs by setting up scaling limits in Cloud Run.
• Use Cloud SQL for database management instead of local storage.
• Consider Google App Engine for simpler deployment with auto-scaling.

3. Deploying on Azure

Azure provides multiple deployment options such as Virtual Machines (VMs), Azure App Service, and Azure Functions. Here, we’ll focus on deploying with Azure App Service.

Using Azure App Service

1. Install Azure CLI and login:

   az login
   az group create --name streamlit-rg --location eastus
   

2. Create an Azure Web App:

   az webapp up --name streamlit-app --resource-group streamlit-rg --runtime "PYTHON:3.9"
   

3. Deploy your code using GitHub Actions or ZIP deployment:

   az webapp deployment source config-zip --resource-group streamlit-rg --name streamlit-app --src app.zip
   

4. Set startup command for Streamlit:

   az webapp config set --resource-group streamlit-rg --name streamlit-app --startup-file "streamlit run app.py --server.port 8000"
   

5. Access your app at https://streamlit-app.azurewebsites.net.

Best Practices for Azure Deployment

• Use Azure Blob Storage to handle static files.
• Set up Auto Scaling to optimize resource utilization.
• Consider using Azure Kubernetes Service (AKS) for better scalability.
• Monitor your app using Azure Monitor for performance tracking.

Wrap up

Each cloud provider offers different ways to deploy a Streamlit app:

• AWS EC2 provides full control over the instance.
• GCP Cloud Run offers a serverless, scalable deployment.
• Azure App Service simplifies deployment with managed hosting.

Choosing the Right Cloud Provider